Interview with Elliot Shmukler, CEO and co-founder of Anomalo

A look at data validation and what it means for businesses

Image: Unsplash

Big data is not only one of the fastest-growing fields right now, but it can also be one that costs businesses lots of money - especially when that data is of poor quality.

While it can be a daunting task, understanding and utilizing data can help your business or department make vital decisions. The problem is, as mentioned above, if your data sets are of poor quality or missing pieces, you're left with an incomplete picture of the issue you are addressing, possibly missing the mark altogether.

For that, data validation continues to grow in importance, as companies that specialize in this can help ensure that your data sets are accurate and that the information gleaned from them is correct.

I had the chance to interview Elliot Shmukler of Anomalo, a data validation platform, to learn more about data validation, what it means for businesses, and why it's important, and also learned a bit about what companies with large data sets can do to use it to its fullest.

You can read the full review below.

Care to introduce yourself and your role at Anomalo?

Certainly. I’m Elliot Shmukler, the Co-founder and CEO of Anomalo. We’re based in the San Francisco Bay Area, and I’ve been leading our small but growing team since our founding in 2018. I’ve had previous roles as a Product and Growth leader at tech companies like Instacart, LinkedIn, and Wealthfront. 

In just a few sentences, what is Anomalo?

Anomalo is a platform that helps companies with data validation - in other words, making sure that the data companies are using to make business decisions and build products is accurate, complete, and consistent with their expectations.

The basic premise here is that, as businesses rely increasingly on data at huge volumes (and often from multiple sources) to make choices on product, marketing, and strategy, they need to make sure that the data they’re using is actually accurate. Otherwise, they can easily make the wrong decisions.

What inspired the creation of the company?

My co-founder Jeremy and I spent many years around Silicon Valley observing first-hand the critical importance of good, high-quality data in making sound business decisions and building great products and services, including at Instacart, where we first met.

While there, we saw a lot of situations where Instacart’s customers had poor experiences due to bad or inaccurate data or where the company made poor decisions because of poor data inputs. So when we left Instacart within about 6 months of each other, and got together to brainstorm some startup ideas, we realized that this data quality and data validation problem we’d seen for years still hadn’t been solved. So, we started Anomalo to bring our best ideas to the problem. 

Why has data validation become so important? Is it the growth of machine learning/algorithmic development, or something else?

Quite simply, companies are using data a lot more to make decisions and manage their products and services. They are also importing and aggregating greater volumes of data from a larger number of disparate sources. 

This creates two issues: with data being used more, the consequences of using inaccurate, corrupted, or stale data are much more serious. And with a greater volume and diversity of data coming in, the problem of actually validating data correctness has also grown a lot more complicated as well. Anomalo helps make the data validation process a lot easier and thus can help companies avoid the cost of making decisions from bad data. 

Machine Learning and AI adoption are also driving this, of course, as those are powerful new technologies that have increased the value and need for data in many companies. But Machine Learning models are also susceptible to the problem of “garbage in, garbage out,” meaning that ML models provide results based on the data they’re ingesting; and if the data is bad, so are the results. This speaks to the importance of data validation tools like Anomalo, which themselves use Machine Learning to make sure that what’s being put into your ML models is as high-quality as you want the result to be. 

Who is Anomalo meant for? Big business, SMBs, or a little of both?

Anomalo is most useful for organizations that have a lot of data, data from a large variety of sources, and use data extensively to manage and grow their businesses. We’ve found a lot of resonance so far among companies in spaces like e-commerce, fintech, social media, and adtech. 

How quickly could a business deploy your tools? Do you help with the setup?

Anomalo can be deployed in under an hour. Sometimes we’ll even book an hour-long Zoom with new customers where we’re able to both deploy and start using the product on their data within the meeting.

Of course, deployment is just the first step. After Anomalo is available, businesses still have to point it at the right data sets and metrics to make sure the right things are being monitored and validated. We help them consistently throughout that process and are available on Slack and Zoom to give as much support as needed.  

Do you have any tips for businesses that are looking to do more with their data?

First off, it’s critical to democratize access to data. Once more people in an organization have access to high-quality data, it will get used more and will add more value. 

Second, every key business area should have a set of clear, reliable, accessible, and continuously updated dashboards to describe how the key metrics in that area are moving and generally what’s going on. This is critical to managing those businesses and empowering teams to understand how their areas are doing and where they can have impact. 

Third, if a business metric changes unexpectedly, it’s important to understand why it changed and what has happened. Very often, such changes result in insights that can highlight opportunities to improve the business or problems that need to be solved if the business is to perform. Without taking the time to develop such understanding (and tools like Anomalo and others can help here), it’s very easy to miss these opportunities to improve.